A novel risk stratification approach and molecular subgroup characterization based on coagulation related genes in colon adenocarcinoma

Abstract

Colon adenocarcinoma (COAD) represents a significant health concern within the population. Advancing our understanding of COAD is imperative for early detection, enabling personalized treatment interventions, and facilitating the development of effective preventive measures. The coagulation system plays a role in tumor-related pathological processes; however, its specific involvement in COAD and potential contributors remain unclear. This study aimed to establish a novel risk stratification approach by analyzing coagulation related genes (CRGs) associated with COAD. Through a comprehensive bioinformatics analysis of data from public databases, we screened COAD associated CRGs and characterized the associated molecular subtypes. After a comprehensive analysis of the characteristics of each subtype, we applied differentially expressed genes in CRG subtypes to establish a new risk stratification method. Clinical subgroup analysis, immunoinfiltration analysis, therapeutic reactivity prediction and other analytical methods suggest the potential clinical value of the established risk stratification method. As one of the selected targets, the effect of MS4A4A on the proliferation and invasion of COAD was confirmed by in vitro experiments, which partially verified the reliability of bioinformatics results. Our findings delineate CRGs potentially implicated in COAD pathogenesis and offer fresh insights into the influence of the coagulation process on tumorigenesis and progression.

Keywords: Coagulation related genes; Colon adenocarcinoma; Immune infiltration; MS4A4A; Risk stratification.

Fig. 1

Analysis of differential expression and genetic variation features of coagulation-related genes (CRG) in COAD. (A) Differential expression analysis of CRG in normal and COAD samples. (B) Assessment of mutation features of CRG gene signatures. (C) Quantitative assessment of copy number variation frequency in COAD. (D) Analysis of prognosis of CRG gene signatures in COAD. (E) PPI network diagram revealing potential interaction relationships among CRG gene signatures

Fig. 2

Molecular subtype identification, mutation characterization and prognosis analysis based on CRG gene signatures in COAD. (A) PCA pattern recognition plot of CRG molecular subtypes. (B) Evaluation of clinical overall survival (OS) rates in CRG molecular subgroups. (C) Prediction analysis of KEGG signaling pathways between CRG molecular subtypes. (D) Tumor mutation characterization of CRG subtypes

Fig. 3

Evaluation of immune infiltration characteristics and immune functional phenotypes in CRG molecular subtypes. (A-D) Assessment of immune status in CRG molecular subtypes. (E) Quantitative analysis of relative proportions of immune cells and immune functional phenotypes of CRG molecular subtypes. (F) Phenotypic assessment of response to CTLA-4/PD-1 immune therapy in CRG molecular subtypes

Fig. 4

Gene subtype identification based on CRG subtype-related genes. (A) Analysis of differential gene expression between CRG subtypes, with filtering threshold: |FC|>2, p.adjust < 0.05. (B, C) GO and KEGG enrichment analysis predictions of differentially expressed genes in CRG subtypes. (D) Gene subtype identification analysis based on the expression profile of differentially expressed genes in CRG subtypes. (E) Differential analysis of CRG gene signatures in gene subtypes

Fig. 5

Construction of the CRG scoring system and stability evaluation for predicting clinical prognosis in COAD. (A, B) Identification of gene signatures associated with COAD prognosis based on LASSO-univariate Cox analysis. (C, D) Model construction and time-dependent ROC curve analysis of the CRG scoring system in the complete cohort. (E-H) Establishment of the CRG scoring system model and time-dependent ROC curve analysis in the training and validation cohorts. (I) Sankey diagram analysis of the relationships among molecular subtypes, CRG scoring system, and clinical prognosis. (J, K) Differential analysis of CRG scores in CRG subtypes and gene subtypes

Fig. 6

Comprehensive Evaluation of the Independent Prognostic Value of the CRG Scoring System and Establishment of the Nomogram Model. (A-E) Distribution of CRG scoring across different clinical-pathological features of COAD. (F, G) Univariate and multivariate Cox analyses of clinical-pathological variables and the CRG scoring system. (H) Establishment of the nomogram model based on clinical variables and CRG scoring. (I) Decision curve analysis (DCA) curve computation. (J) Concordance index evaluation of different variable features at different survival times. (K) Analysis of consistency index between predicted OS using the nomogram model and actual OS over time

Fig. 7

Correlation analysis of the CRG scoring system with immune infiltration microenvironment and somatic mutation characteristics. (A-D) Evaluation of the immune infiltration status of the CRG scoring system. (E) Quantitative analysis of immune cell proportions and immune function scores in CRG scoring subgroups. (F) Assessment of somatic mutation characteristics in CRG scoring subgroups

Fig. 8

Prediction of the response of CRG scoring subgroups to PD-1/CTLA-4 and targeted drugs. (A-D) Prediction of the response to immune therapy based on the TCIA database. (E-L) Assessment of potential targeted drug sensitivity associated with CRG scoring subgroups

Fig. 9

Effects of MS4A4A on cell proliferation and migration in COAD cell lines. (A) Protein expression levels of MS4A4A across various COAD cell lines. (B) Representative western blot bands illustrating MS4A4A overexpression in HCT116 cells. (C) Representative bands depicting MS4A4A knockdown in Caco2 cells. (D) RT-PCR results demonstrating the efficiency of MS4A4A overexpression in HCT116 cells. (E) RT-PCR results showing the efficiency of MS4A4A knockdown in Caco2 cells. (F, G, H) CCK8 assay and clone formation assay conducted in Caco2 cells with low MS4A4A expression. (I, J) Transwell assay carried out in HCT116 cells with stable MS4A4A overexpression